pandas：根据一列中的值计算出的列 [英] pandas: Calculated column based on values in one column

问题描述

我在csv文件中有这样的列（我使用 read_csv（'fileA.csv'，parse_dates = ['ProcessA_Timestamp']）加载

I have columns like this in a csv file (I load it using read_csv('fileA.csv', parse_dates=['ProcessA_Timestamp']))

Item    ProcessA_Timestamp
'A'    2014-06-08 03:32:20
'B'    2014-06-08 03:32:20
'A'    2014-06-08 03:33:19
'C'    2014-06-08 03:33:20
'B'    2014-06-08 03:33:40
'D'    2014-06-08 03:38:20

我将如何创建名为 ProcessA_ProcessingTime 的列，这将是上次时间与表<$ c $中出现时间之间的时间差c>- 第一次出现在表格中。

How would I go about creating a column called ProcessA_ProcessingTime, which would be the time difference between last time an item occurs in the table - first time it occurs in the table.

类似地，我还有其他数据框（ （不确定是否应将它们合并到一个数据帧中。）具有自己的 Process * _Timestamp s的数据框。

Similarly, I have other data frames (which I'm not sure if they should be merged into one dataframe).. that have their own Process*_Timestamps.

最后，我需要创建一个表，其数据如下所示：

Finally, I need to create a table, where the data is like this:

Item ProcessA_ProcessingTime ProcessB_ProcessingTime ... ProcessX_ProcessingTime
'A'                 00:00:59                  ...
'B'                 00:01:21
'C'         NOT FINISHED YET
'D'         NOT FINISHED YET

推荐答案

使用pandas groupby-apply组合。按项目对数据框进行分组，并应用一个函数来计算处理时间。

You can use the pandas groupby-apply combo. Group the dataframe by "Item" and apply a function that calculates the process time. Something like:

import pandas as pd

def calc_process_time(row):
    ts = row["ProcessA_Timestamp].values
    if len(ts) == 1:
        return pd.NaT
    else:
        return ts[-1] - ts[0] #last time - first time

df.groupby("Item").apply(calc_process_time)

这篇关于pandas：根据一列中的值计算出的列的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！